| Age | Commit message (Collapse) | Author | 
|---|
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  | to stop at 5000 nodes to keep from normalizing 'forever' | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  | slipped through; testdata/scad/misc/normal-nan.scad | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  | without eigen2 | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  |  | 
|  | transformed correctly | 
|  |  | 
|  |  | 
|  | to output the expected image | 
|  | fixing intermittent crash bug | 
|  |  | 
|  | objects which don't touch won't be visible in thrown together mode. Use debug operators in such cases | 
|  | of difficult models in OpenCSG preview mode | 
|  |  | 
|  |  | 
|  | components with very different orders of magnitude, as may happen when floating point errors fail to produce a 0) | 
|  |  | 
|  | Rewrote the algorithm based on the original Goldfeather paper. Fixes #44 | 
|  |  |